South Middlesex Times

Researchers at Rutgers University-New Brunswick have developed an artificial intelligence (AI) tool to predict endangered whale habitats, aiming to guide ships along the Atlantic coast to avoid collisions. This initiative is intended to prevent accidents and support conservation strategies and responsible ocean development.

The AI program improves current methods of monitoring marine species distribution, including the critically endangered North Atlantic right whale. Listed as endangered since 1970 under the Endangered Species Act, there are approximately 370 individuals remaining, with about 70 reproductively active females, according to the U.S. National Oceanic and Atmospheric Administration.

Ahmed Aziz Ezzat, an assistant professor in the Department of Industrial and Systems Engineering, led the effort alongside Josh Kohut, a professor in marine sciences and dean of research at the School of Environmental and Biological Sciences. Jiaxiang Ji, a doctoral student in the School of Engineering, significantly contributed to the project. Their report was published in Nature Scientific Reports.

"This is a demonstration of the power of employing AI methodologies to advance our ability to predict or estimate where these whales are," said Kohut.

Kohut compared the program's output to tracking people's movements in a house based on environmental factors like food availability or television activity. "With this program, we’re correlating the position of a whale in the ocean with environmental conditions," he explained. "This allows us to become much more informed on decision making about where the whales might be."

Initially focused on developing models for North Atlantic right whale presence near offshore wind farms, researchers found broader implications for their work. "These tools are valuable and would solidly benefit anyone engaged in the blue economy – including fishing, shipping and developing alternative forms of energy sustainably," Ezzat stated.

The machine-learning program analyzes large data sets for patterns and relationships without explicit instructions. "The outcome...is basically a prediction of where and when you will have a higher likelihood of encountering a marine mammal," Ezzat described as a "probability map."

Data analyzed by the model includes underwater glider and satellite-based data from Rutgers University Center for Ocean Observing Leadership since 1992 and satellite data from the University of Delaware. Underwater gliders measure seawater aspects such as temperature and salinity while recording underwater calls from marine mammals.

"We’ve had the data but, until now, we’ve not been able to put...those detections...and what the environment is like at those places – together," Kohut said.

Other contributors included Laura Nazzaro from Marine and Coastal Sciences; Jeeva Ramasamy, an undergraduate computer science major; among others involved in this study.